Post on 12-Jun-2020
transcript
North America Dynamic WindNorth America Dynamic WindNorth America Dynamic Wind North America Dynamic Wind Generator Modeling UpdateGenerator Modeling Update
Based on work performed by the WECC Wind Generator Modeling Group and the IEEE Dynamic Performance of Wind Power Generation Working Group
Abraham Ellis
Sandia National Laboratories
Yuriy Kazachkov
Siemens‐PTI
US Membership to the IEC TF88 WG27
aellis@sandia.gov
Eduard Muljadi
National Renewable Energy Laboratory
yuriy.kazachkov@siemens.com
Juan Sanchez‐Gasca, Nicholas Miller
GE Energy ‐ PSLF
eduard.muljadi@nrel.gov
Robert Zavadil
EnerNex Corporation
bobz@enernex.com
juan1.sanchez@ge.com, nicholas.miller@ge.com
Pouyan Pourbeik
Electric power Research Institute
PPourbeik@epri.combobz@enernex.com PPourbeik@epri.com
Roskilde, Denmark – October 2009
Presentation ContentsPresentation ContentsPresentation ContentsPresentation Contents
1. Status of WECC WGMG generic d li ff t d l t d ti itimodeling effort and related activities
in IEEE & NERC.
2 Review of WECC generic model2. Review of WECC generic model structures, testing and verification
3 Simplified aerodynamic conversion3. Simplified aerodynamic conversion representation for generic models
4. Model validation and parameter pidentification
5. Wind power plant network equivalent for power flow and dynamic simulations
Wind Turbine Generator TopologiesWind Turbine Generator TopologiesWind Turbine Generator TopologiesWind Turbine Generator Topologies
• Four basic WTG types Type 1 Type 2based on grid interface technology
Type 1 Fixed speed conventional
generator
PlantFeeders
PF controlac
generator
PlantFeeders
PF control– Type 1 – Fixed-speed, conventional induction generator
– Type 2 – Variable slip, induction generators with variable rotor
Slip poweras heat loss
capacitorstodc
capacitors
generators with variable rotor resistance
– Type 3 – Variable speed, doubly-fed asynchronous generators with rotor
PlantFeeders
ac dcgenerator
PlantFeeders
Type 3 Type 4
asynchronous generators with rotor-side converter
– Type 4 – Variable speed, asynchronous generators with full
generator
full power
actodc
dctoacac
todc
dctoac
asynchronous generators with full converter interface
partial power
WECC Generic Model ImplementationWECC Generic Model ImplementationWECC Generic Model ImplementationWECC Generic Model Implementation• Completed generic models implemented as standard‐library
models in PSSE/PSLF completedmodels in PSSE/PSLF completed
PSLF/17 Model Type Type 1 Type 2 Type 3 Type 4Generator wt1g wt2g wt3g wt4gExcitation / Controller wt2e wt3e wt4eTurbine wt1t wt2t wt3t
Generic model WT1 WT2 WT3 WT4Generator WT1G WT2G WT3G WT4GEl. Controller WT2E WT3E WT4ETurbine/shaft WT12T WT12T WT3T
PSSE/32
Pitch Controller/Pseudo Gov. wt1p wt2p wt3p
• Key ParticipantsSiemens (PSSE) General Electric (PSLF) DOE (Sandia National
Pitch control WT3PPseudo Gov/: aerodynamics WT12A WT12A
– Siemens (PSSE), General Electric (PSLF), DOE (Sandia National Laboratories and NREL), Consultants, Universities, other stakeholders
• Current focus– Additional model validation, refinement, upgrade
– Identification of generic model parameters
Partial List of WTG TypesPartial List of WTG Types
Type 1 Type 2 Type 3 Type 4 V t NM72 V t V80 GE GE 2 5XL
Partial List of WTG TypesPartial List of WTG Types
Vestas NM721.65 MW, 50/60 Hz
Vestas V801.8 MW, 60Hz
GE 1.5 MW, 50/60 Hz
GE 2.5XL2.0 MW, 50/60 Hz
Vestas V82 1.65 MW, 50/60Hz
Vestas V47 660 kW, 50/60 Hz
GE 3.6 MW, 50/60 Hz
Clipper 2.5 MW, 50/60 Hz
BONUS Gamesa G80 Gamesa G80 Enercon E66 (now Siemens)
1.3 MW, 50/60 Hz 1.8 MW, 60 Hz 2 MW, 50 Hz 1.8 MW, 50 Hz
BONUS (now Siemens) 2.3 MW, 50 Hz
Suzlon S88 2.1 MW, 50Hz
NORDEX N80 2.5 MW, 50Hz
Enercon E70 2.0 MW, 50 Hz
Mitsubishi MWT100a1 MW, 60 Hz
REPower MD70 and MD77
1.5 MW, 50Hz
Siemens, 2.3VS82, 2.3MW, 50/60Hz
Suzlon S66 1.25 MW, 50 Hz
REPower MM70 and MM82 2.0
Kennetech 33-MVS, 400kW, 60Hz1.25 MW, 50 Hz MM70 and MM82 2.0
MW, 50Hz 400kW, 60Hz
Mitsubishi MWT-92/95
2.4 MW
Fuhrlaender FL 2 5 Fuhrlaender FL 2.5 MW, 60 Hz
Related ActivitiesRelated ActivitiesRelated ActivitiesRelated Activities• IEEE DPWPG Task Force
V ifi ti h t f WECC i d l– Verification, enhancement of WECC generic models
– Paper on Specifications for Wind Power Plant models
– Paper on Validation Procedures for Wind Power Plant modelsPaper on Validation Procedures for Wind Power Plant models
– Outreach (Tutorials)
– Coordination with WECC, NERC (IVGTF), UWIG, CIGRE, IEC, …, ( ), , , ,
• UWIG/EnerNex (DOE funding)D l d t ti f i d t bi ifi– Develop user documentation for generic and turbine‐specific wind turbine models; conduct training seminars
– Verify performance of generic and vendor‐specific modelsy p g p
– Validate of models against field test results
Related ActivitiesRelated ActivitiesRelated ActivitiesRelated Activities
• NERC IVGTF Work Plan, Task 1.1 (Planning)– Valid, generic, non‐confidential, and public standard power flow and stability (positive‐sequence) models […] needed
– Models should be readily validated and publicly available to– Models should be readily validated and publicly available to power utilities and all other industry stakeholders
– Model parameters should be provided by manufacturers
– Common model validation standard […] should be adopted
– Review the Modeling, Data and Analysis (MOD) standards t id tif d difi ti t dd i blto identify any need modifications to address variable generation modeling and model validation
Good Progress, More Work NeededGood Progress, More Work NeededGood Progress, More Work NeededGood Progress, More Work Needed• Generic models emerging as the correct approach
– Pressure continues to build in North America, Europe, p
– Several technical activities underway to validate, improve
– Strong trend among manufacturers to replace custom models with adjusted generic modelsadjusted generic models
• Need to build on recent progress with generic models– Include a more complete set of control options
• Programmed inertia, frequency support features, LVRT details, etc.
– Perform model validation, develop parameter sets
– Develop model documentation and application guides
• Coordination of related efforts is key
• Ultimate goal: full‐featured, industry standard models– In the meantime, manufacture models are still useful in cases where
specific control features not available in the generic models have a substantial impact
D i ti f WECC G i M d lD i ti f WECC G i M d lDescription of WECC Generic ModelsDescription of WECC Generic Models
References:– Y. Kazachkov, S. Stapleton, “Do Generic Dynamic Simulation Wind
Turbine Models Exist?”, WindPower 2005, Denver, Colorado, May 2005
– N. W. Miller, W. W. Price, J. J. Sanchez‐Gasca, “Dynamic Model of GE’s 1.5 and 3.6 MW Wind Turbine Generators – Model Structure, Simulation Results, and Model Validation”, CIGRE Technical Brochure 328, Modeling and Dynamic Performance of Wind Generation as it R l P S C l d D i P f CIGRE WGRelates to Power System Control and Dynamic Performance, CIGRE WG C4.601, August, 2007
– WECC WGMG, “Generic Wind Plant Models for Power System Studies”, Wi dP 2006 Pi b h PA J 2006WindPower 2006, Pittsburgh, PA, June 2006
– WECC WGMG, “Development and Validation of WECC Variable Speed Wind Turbine Dynamic Models for Grid Integration Studies”, Wi dP 2007 L A l CA J 2007WindPower 2007, Los Angeles, CA, June 2007
WECC Generic ModelsWECC Generic Models
• WECC WGMG GoalsS if i i t d l f l l– Specify generic, non‐proprietary models for large‐scale power system simulations (positive‐sequence models)
– Generic models are parametrically adjustable to any specific p y j y pwind turbine of the same type in the market
– Models are simplified, but major dynamic behavior is maintained • Find balance between simplification and performance For example• Find balance between simplification and performance. For example,
representation of aerodynamic conversion should be simplified to avoid using proprietary Cp curves. Too much simplification is not good: a model does not perform well if aerodynamics is ignored (e.g. constant mechanical power)
– Perform model testing and as much validation as possible
Implement in PSSE and PSLF with full documentation and– Implement in PSSE and PSLF with full documentation and default parameter sets
WECC Generic ModelsWECC Generic Models
• Technical SpecificationsApplication: electrical disturbances (not wind disturbances) primarily– Application: electrical disturbances (not wind disturbances), primarily grid faults external to the plant, typically 3 to 6 cycle duration.
– Typical simulation time frame of interest are 20 to 30 seconds, with a ¼ cycle integration time step Wind speed assumed to be constantcycle integration time step. Wind speed assumed to be constant.
– Model able to handle oscillatory modes from dc to 5 Hz.
– Initialize from power flow at full or partial power; able to handle user‐ifi d i d dspecified wind speed
– Speed and voltage protection modeled separately
– Represent machine inertia and first shaft torsional mode characteristics
– The models should be applicable to strong and weak systems with a short circuit ratio of 2.5 and higher at the point of interconnection
– Shunt capacitors and any other reactive support equipment modeled p y pp q pseparately with existing standard models
Response to Frequency DisturbancesResponse to Frequency Disturbances
• The generic WTG models were not developed with g pthe intent of being accurate for the study of frequency excursions on the power system, or to reproduce the behavior of advanced powerreproduce the behavior of advanced power management features that are imminently becoming available from some WTG manufacturers (such as programmed inertia and 'spinning reserve' by spilling wind). W d t k f th ith M f t t• We need to work further with Manufacturers to better understand response to frequency disturbances and thus improve the models.p
WT1 Generic ModelWT1 Generic ModelWT1 Generic ModelWT1 Generic ModelBasic Description
Modules: WT1T, WT1P, WT1G
Generator model is conventional induction machine
Default model data provided in documentation (Mitsubishi MWT1000A)
Single or two-mass shaft model
PlantFeeders
S g e o t o ass s a t ode
Power factor correction capacitor must be provided and initialized from load flow data
generator
Feeders
PF controlcapacitors
No special initialization or run-time script needed
No special adjustment needed for wind farm representationp
WT1 Generic ModelWT1 Generic ModelWT1 Generic ModelWT1 Generic Model
WT1G 2lr
2ro
LRω
2lrL.12qrψ-
+
+
-
SLIPωo2drψ
Induction Generator Model
1lr
1ro
LRω
1lrL.1
1qrψ-
++
- +
+
satm "Lqd "ψ"E =−
iqsmqψ +oω
SLIPωo1drψ
mq2
md2 ψψ +
ids
satm "Liqsmqψ
-
-mψ
Se )ψ(S.1K mesat +=
2lr1lrmsatsatm L/.1L/.1L/K
.1"L++
=
mdψ
satm "L
SLIPωo
1lr
1ro
LRω
1lrL.1
1qrψ
1drψ
+
+
-sat"Lm
dq "ψ"E =
+
+
+
LLL
satmL
+
SLIPωo
2lr
2ro
LRω
2qrψ
2drψ
+
+
-
+
)Rω(/)'LL()"LRω(/'LL"T
)Rω(/)LL()'LRω(/LL'TL"L)L/.1L/.1L/.1(/.1"L
L'L)L/.1L/.1(/.1'LLLL
2rom2lrm2rom2lro
1rom1lrm1rom1lro
l2lr1lrmm
l1lrmm
lsm
+==
+==
−=++=
−=+=
−=
2lrL.1
Model simulates a one-cage model is used if Lpp is equal to Lp. Otherwise, it simulates a two-cage induction generator. If Lpp or Tppo are 0., the model sets Lpp = Lp.
WT1 Generic ModelWT1 Generic ModelWT1 Generic ModelWT1 Generic Model
WT1PPitch Control Model
speed
Kp iFrom
Kdroop1
Σ
Σ1 + sT
Kp
Ki 1s
Σ
pimax
pimin
11 + sT1
11 + sT2
pgen
pmechwref
s1
s2 s3To
GovernorModel
TurbineModel
p
s0
1 + sTpe
pref
s1From
GeneratorModel
Note: For disturbances involving frequency drops this is not necessarily accurate – should set pimax to rated turbine output to a oid an response from the t rbineavoid any response from the turbine.
WT1 Generic ModelWT1 Generic ModelWT1 Generic ModelWT1 Generic ModelHt = Htfrac H
Hg = H - Ht
K = 2 (2π Freq1 ) H t2
H
HgWT1T Two-mass model
Tmech 1s
12Ht
Σ
++ -
-ωο
+
+ω tΣ
δ
Pmech
ω t
Δωt
Δ Δω
..s0
FromGovernor
Model
Two mass model
K
1s
12Hg
ΣDshaft
Σ
+
+Telec
-
ωο+
+ωgΣ-
1s
δtg
Pgen Δωg
Δω tg Δω tg
..s2
s1
FromGenerator
Model
FromGenerator
Model
οωg
Pmech ω1s
12H+
Pgen
TaccFromGovernor
ModelΣ
..To
Generators0
WT1T Single mass model (Htfrac = 0)
D
GeneratorModel andGovernor
Model
WT1 Fault TestWT1 Fault TestWT1 Fault TestWT1 Fault Test9-cycle, 3-phase fault at POI, clear fault by tripping line
V1
Pg
V5V5
Speed
Qg
WT1 Fault TestWT1 Fault TestWT1 Fault TestWT1 Fault Test9-cycle, 3-phase fault at POI, Self-Clearing
V1
Pg
V5
Qg
Speed
WT1 Frequency Drop TestWT1 Frequency Drop TestWT1 Frequency Drop TestWT1 Frequency Drop Test1% drop at t = 1 sec
V1
V5
Pg
QgSpeed Qgp
Frequency at Bus 5
WT1 Model ComparisonWT1 Model ComparisonWT1 Model ComparisonWT1 Model ComparisonAgainst Mitsubishi MWT1000A Manufacturer Model
WT2 Generic ModelWT2 Generic ModelWT2 Generic ModelWT2 Generic Model
Basic Description
Modules: WT2T, WT2E, WT2A/WT2P, WT2G
Generator model is conventional induction machine
Default model data provided in documentation (Vestas V80)( estas 80)
Single or two-mass shaft model
Power factor correction capacitor must be provided p pand initialized from load flow data
No special initialization or run-time script neededgenerator
PlantFeeders
PF controlac No special adjustment needed for wind farm representationSlip power
as heat loss
PF controlcapacitors
actodc
WT2 Generic ModelWT2 Generic ModelWT2 Generic ModelWT2 Generic Model
PgenFrom
Σ
Rmax
+
-
genFromGenerator
Model
T
Rext
Kp
1 + sTp
Kpp + Kip / s
s0
s1
WT2E
1 + sTw
+
Δω
P vs. slip curve
SpeedFrom
TurbineModel
ToGenerator
ModelRmin
s2
Kw
speed
Kp pimaxFrom
Turbine
WT2P
Kdroop1
Σ
Σ1 + sT
p
Ki 1s
Σ
pimax
pimin
11 + sT1
11 + sT2
pgen
pmechwref
1
s2 s3To
GovernorModel
TurbineModel
p
s0
1 + sTpe
pref
s1From
GeneratorModel
WT2 Generic ModelWT2 Generic ModelWT2 Generic ModelWT2 Generic ModelHt = Htfrac H
Hg = H - Ht
K = 2 (2π Freq1 ) H t2
H
HgWT2T Two-mass model
Tmech 1s
12Ht
Σ
++ -
-ωο
+
+ω tΣ
δ
Pmech
ω t
Δωt
Δ Δω
..s0
FromGovernor
Model
Two mass model
K
1s
12Hg
ΣDshaft
Σ
+
+Telec
-
ωο+
+ωgΣ-
1s
δtg
Pgen Δωg
Δω tg Δω tg
..s2
s1
FromGenerator
Model
FromGenerator
Model
οωg
Pmech ω1s
12H+
Pgen
TaccFromGovernor
ModelΣ
..To
Generators0
WT2T Single mass model (Htfrac = 0)
D
GeneratorModel andGovernor
Model
WT2 Fault TestWT2 Fault TestWT2 Fault TestWT2 Fault Test9-cycle, 3-phase fault at POI, clear fault by tripping line
P
V1
V5 Pg
Qg
Speed
WT2 Fault TestWT2 Fault TestWT2 Fault TestWT2 Fault Test9-cycle, 3-phase fault at POI, self-clearing
V1
V
Pg
V5 Qg
Speed
WT2 Frequency Drop TestWT2 Frequency Drop TestWT2 Frequency Drop TestWT2 Frequency Drop Test1% drop at t = 1 sec
V5
V1
SpeedPg
Speed
QgFrequency at Bus 5
WT2 Model ComparisonWT2 Model ComparisonWT2 Model ComparisonWT2 Model ComparisonAgainst Vestas V80 Manufacturer Model
WT3WT3 GenericGeneric ModelModelWT3 WT3 Generic Generic ModelModel
Vreg bus V Basic Description
Generator/Converter
M d l
ConverterControlM d l
Ip (P)Command
g Vterm
Eq (Q)Command
Basic Description
Modules: WT3T, WT3E, WT3P, WT3G
Generator is modeled as controlled currentModelModel
PowerOrder
Pgen , Qgen
ShaftSpeedSpeed
Order
Pgen , Qgen
Pgen
Generator is modeled as controlled current injection
Default model data provided in documentation (GE 1 5)
Pitch ControlModel
WindTurbineModel
BladePitch
(GE 1.5)
Single or two-mass shaft model
No special initialization or run-time script neededNo special initialization or run-time script needed
Plant level reactive control options available for wind farm model (PF control, Q control and voltage control)
generator
PlantFeeders
acto
dcto
28
voltage control)
partia l power
todc
toac
WT3 GenericWT3 Generic ModelModelWT3 Generic WT3 Generic ModelModelHt = Htfrac H
Hg = H - Ht
K = 2 (2π Freq1 ) H t2
H
HgWT3T Two-mass model
Tmech 1s
12Ht
Σ
++ -
-ωο
+
+ω tΣ
1 δtg
Pmech
ω t
Δωt
Δω tg Δω tg
..
Two mass model
K
1s
12Hg
ΣDshaft
Σ
+
+Telec
-
ωο+
+ωgΣ-
1s
tg
Pgen
ωg
Δωg
tg tg
..
g
WT3T Single mass model (Htfrac = 0)
29
WT3WT3 Generic ModelGeneric ModelWT3 WT3 Generic ModelGeneric Model
WT3P BladeωF
PImax
θ
Anti-windup onPitch Limits rate lim it (PIrate)
Pitch Control Model θPitch
ωreffFrom
Converter
ωerrω
FromTurbineM odel To
TurbineM odel
11+ sT p
cm dθΣ+
+Pitch
Control
Kpp + Kip / s
Anti-windup onPitch Limits
+
Σ+
PImin
Pord
+Converter
ControlM odel
Pi tchCompensation
K pc+ K ic / sΣ
1Pset
NOTE: Pset should normally be 1.0 unless it is controlled by a separate active power control model, e.g. to provide
i It t l b t thgoverning response. It must always be greater than or equal to the initial power output of the WTG.
30
WT3WT3 Generic ModelGeneric ModelWT3 WT3 Generic ModelGeneric ModelWT3EReactive Power
Wind Plant Reactive Power Control Emulation
Ki / s
VrfqQmax
Control Model Vc
Kiv / s+
11+ sTc
11+ sTr
Qmin
1/FnKpv
1+ sTv
Qwv
+
+
s2
s3 s5
Σ Σ
Pgen 11+ sTp
-1
varflg1
PFAref tan
x Qord
Power Factor Q
Qmax
p
Qref
0
QgenVmax
Vterm XIQmax
E
Power FactorRegulator
QcmdQmin
vltflg
s6
++
Vref Kqv / s
Vmin
Eq cmd
ToGenerator /Converter
Model
Kqi / sQcmd
XIQmin
0
1
s0 s1ΣΣ
31
WT3WT3 Generic ModelGeneric ModelWT3 WT3 Generic ModelGeneric Model
Anti-windupon
(shaft speed)ωWT3E
Active Power (Torque)
Ip cmdTo
Generator /Converter
Model
XPord1
1+ sTpc
Pmax & dPmax/dton
Power Limits
Kptrq+ Kitrq / sωerr1
1 + Tsps
ωref
f ( Pgen ) Σ
+Pgen
Pmin & -dPmax/dt
.
.
Ipmax
( q )Control Model
VtermTo PitchControlModel
To PitchControlModel
0.8
1.0
1.2
(p.u
.)
0.0
0.2
0.4
0.6
Pow
er (
32
0.00.60 0.70 0.80 0.90 1.00 1.10 1.20 1.30
Turbine Speed Setpoint (p.u.)
WT3WT3 Generic ModelGeneric ModelWT3 WT3 Generic ModelGeneric Model
WT3G IE "WT3GGenerator / Converter Model From
exwtge
High VoltageReactive Current
Management
Low Voltage
Isorc-1X"(efd)
11+ 0.02s
s0
Eq"cmd
LVPL & rrpwr
Fromexwtge
gActive CurrentManagement
IPcmd
(ladifd)1
1+ 0.02ss1
IPlv
Vterm
jX"11+ 0.02s
LVPL
LVPL
1.11
V
Lvplsw = 1
Lvplsw = 0
Low Voltage Power Logic
Vbrkpt(0.90)
zerox(0.50)
s2
33
WT3 Fault TestWT3 Fault TestWT3 Fault TestWT3 Fault Test9-cycle, 3-phase fault at POI, clear fault by tripping line
V1
Pg
V5 SQg
V5 Speed
WT3 Fault TestWT3 Fault TestWT3 Fault TestWT3 Fault Test9-cycle, 3-phase fault at POI, self clearing
V1
Pg
V5
Qg
Speed
WT3 Frequency Drop TestWT3 Frequency Drop TestWT3 Frequency Drop TestWT3 Frequency Drop Test1% drop at t = 1 sec
P
V1
V5 Pg5
Speed
QgFrequency at Bus 5
WT3 Model ComparisonWT3 Model Comparison
S ll S t
WT3 Model ComparisonWT3 Model Comparison
L S t
Against GE1.5 Manufacturer Model
Small System Large System
WT4 Generic ModelWT4 Generic ModelWT4 Generic ModelWT4 Generic Model
Basic Description
Modules: WT4T, WT4E, WT4G
No generator model. Just converter represented t ll d t i j tias controlled current injection
Default model data provided in documentation (GE 2.x)
PlantFeeders
Machine dynamics not modeled (it decoupled from the grid by the converter)
No special initialization or run time scriptgenerator
Feedersactodc
dctoac
No special initialization or run-time script needed
Plant level reactive control options available for wind farm model (PF control Q control andfull power wind farm model (PF control, Q control and voltage control)
WT4 Generic ModelWT4 Generic ModelWT4 Generic ModelWT4 Generic Model
WT4GGenerator/Converter Model
Fromt4
High VoltageReactive Current
Management
Isorc
(efd)-1
1+ 0.02s
s0
I Qcmd
wt4e
From
g
Low VoltageActive CurrentManagement
IPcmd
(ladifd)1
1+ 0.02s
s1
LVPL & rrpwr
IPlv
wt4t
Vterm
1LVPL
LVPL
VLvplsw = 0 Lvpl1
jX"11+ 0.02s
Low Voltage Power Logic
Vbrkpt(0.90)
zerox(0.50)
s2Lvplsw = 1
WT4 Generic ModelWT4 Generic ModelWT4 Generic ModelWT4 Generic Model
WT4E Vreg
Kiv/s+
Vrfq(vref)
11+ sTc
11+ sTr
Qmax
1/fN
Qords4
-K
Qwv
+
+ΣΣ
Converter Electrical Control Model
WindCONTROLEmulator
Qmins5s3
Kpv1+ sTv
s2
wv
Qcmd
Qmin
Qmax
Qref(vref) 0
varflg
1
PFAref tan(vref)
+
-1
Pelec
min
0
pfaflg
1
( )tan
x
s6
(vref)
11+ sTp
1
Qord(vref)
Qcmd
Qgen
Vref
Vmax
K i / sIQcmd
K i / s
-Iqmx
ΣΣ++
Vterm
Kvi / s
Vmins0 s1
Kqi / s
-Iqmn
(efd)to Wind
GeneratorModelwt4gConverter Current LimitP,Q Priority Flag
Σ
Pord
Vterm
Ipmx..
IPcmd
(ladifd)to Wind
Generator Modelwt4g
Porx
(vsig)from Wind
Turbine Modelwt4t
WT4 Generic ModelWT4 Generic ModelWT4 Generic ModelWT4 Generic Model
WT4E P,Q Priority Flag
(pqflag)
Converter CurrentLimiter Model
IqmnIqmx
10
Iqmn Iqmx
Vt
-1
P PriorityQ Priority
-1
Iqhl Minimum
Vt
Iqmxv
Iqmxv
1.6qmax
1.0
ImaxTDImaxTD
2 - IPcmd2
Minimum
IPcmd
Minimum
q
ImaxTD2 - IQcmd
2IQcmd
Minimum
Iphl
Minimum
IpmxIpmx
MinimumMinimum
WT4 Generic ModelWT4 Generic ModelWT4 Generic ModelWT4 Generic Model
WT4T
dP PP
Turbine Model
Σ1 + sTpw
1Kpp +
s
Kip
dPmx
+-
-Σ
+
-Pord
PrefPref
Pelec
From Wind To wt4e
piin piou
s0 s1
dPmn
sKf
Generator Modelwt4g(pelec)
To wt4e(vsig)
s2
1 + sTf
WT4 Fault TestWT4 Fault TestWT4 Fault TestWT4 Fault Test9-cycle, 3-phase fault at POI, clear fault by tripping line
V1
VIp
V5 Pg
Qg
Iq
WT4 Fault TestWT4 Fault TestWT4 Fault TestWT4 Fault Test9-cycle, 3-phase fault at POI, clear fault by tripping line
V1
P
Ip
Pg
Qg
V5
Iq
WT4 Frequency Drop TestWT4 Frequency Drop TestWT4 Frequency Drop TestWT4 Frequency Drop Test1% drop at t = 1 sec
V1
Qg
V5
Pg
FrequencyFrequency at Bus 5
WT4 Model ComparisonWT4 Model ComparisonWT4 Model ComparisonWT4 Model ComparisonAgainst GE 2.5 MW Manufacturer Model
Terminal VoltageTerminal Voltage
PelecPelec
QelecQelec
At T=1.0 sec., place fault at At T=1.0 sec., place fault at POI, clear in 250 msPOI, clear in 250 ms
WT4 Model TestingWT4 Model TestingWT4 Model TestingWT4 Model TestingAgainst GE 2.5 MW Manufacturer Model
WIPCMNDWIPCMND
PelecPelecPelecPelec
QelecQelec
At T=0 1 sec converter current limit wasAt T=0 1 sec converter current limit wasAt T 0.1 sec., converter current limit was At T 0.1 sec., converter current limit was reduced from 1.7 p.u. to 0.8 p.u. and reduced from 1.7 p.u. to 0.8 p.u. and restored back to 1.7 p.u. at T=4.1 sec.restored back to 1.7 p.u. at T=4.1 sec.
WT4 Model VerificationWT4 Model VerificationWT4 Model VerificationWT4 Model VerificationAgainst ABB Converter PSCAD Model
From “WECC - Model Specifications, Validation”, presentation by Slavomir Seman, 7/29/09 at the IEEE PES General Meeting, Calgary, Canada.
WT4 Model VerificationWT4 Model VerificationWT4 Model VerificationWT4 Model VerificationAgainst ABB Converter Full Power Test
From “WECC - Model Specifications, Validation”, presentation by Slavomir Seman, 7/29/09 at the IEEE PES General Meeting, Calgary, Canada.
On Aerodynamic Simplification used in WECC On Aerodynamic Simplification used in WECC Generic Type 3 modelGeneric Type 3 modelypyp
References:– W. W. Price, J.J. Sanchez‐Gasca, “Simplified Wind Turbine Generator Aerodynamic Models for Transient Stability Studies” Proc. IEEE PES PSCE, Atlanta, Georgia, October‐November 2006
– M Behnke A Ellis Y Kazachkov T McCoy E Muljadi WM. Behnke, A. Ellis, Y. Kazachkov, T. McCoy, E. Muljadi, W. Price, J. Sanchez‐Gasca, “Development and Validation of WECC Variable Speed Wind Turbine Dynamic Models for G id I t ti St di ” AWEA Wi dP L A lGrid Integration Studies”, AWEA WindPower, Los Angeles, California, June 2007
Aerodynamic ConversionAerodynamic ConversionAerodynamic ConversionAerodynamic Conversion
Pelec
Pmech ωAerodynamic
Model
W indSpeed Rotor
Modelgen
θ
BladePitch
ω rotor
Pd
Pelec Controlordelec
System
)(3 θλρ CvAhP =
51
),(2
θλpCvrAmechP =
AerodynamicAerodynamic SimplificationSimplificationAerodynamic Aerodynamic SimplificationSimplification
• Assume wind speed is constant for the duration of the typicalAssume wind speed is constant for the duration of the typical period of interest in dynamic simulations (up to about 20 seconds)
• For variable speed Type 3 it was noticed that
– Rate of change of mechanical power (Pmech) varies linearly with respect to pitch angle (θ ) in the range 0 < θ < 30 degwith respect to pitch angle (θ ) in the range 0 < θ < 30 deg
– Pmech varies linearly with respect to wind speed (Vw) from cut‐in to rated wind speed
– θ varies linearly with respect to Vw for wind speeds above rated
52
Aerodynamic Model SimplificationAerodynamic Model SimplificationAerodynamic Model SimplificationAerodynamic Model Simplification
Detailed 3-D Cp CurveDetailed 3-D Cp Curve.
2-D Cp Curve used in detailed models
0.2
0.3
0.4
0.5
Cp
θ=1o
θ=3o
θ=5o
θ=7o
θ=9o
θ=11o
=13o
Linear relationsallow for simplifications
0 2 4 6 8 10 12 14 16 18 20
0
0.1
λ
θ=13o
θ=15o
53
ExampleExample for GEfor GE 1.51.5 WTGWTGExample Example for GE for GE 1.5 1.5 WTGWTG
“Aerodynamic governor” modelAerodynamic governor model
Pm = Pmo – θ ( θ ‐ θ o ) / 100
Initialization:
P P l (f fl )• Pmo = Pelec (from power flow)• Use Fig. 8 to find Vw for θo = 0• Use Fig. 9 to compute θo if
d d iPe = Prated and user‐input Vw is greater than rated wind speed
55
Testing of Aerodynamic SimplificationTesting of Aerodynamic SimplificationTesting of Aerodynamic SimplificationTesting of Aerodynamic SimplificationAgainst GE1.5 Manufacturer Model
At rated output At 50% output
Blue = manufacturer standard model, Red = WECC WT3 generic model
On Identification of WTG Model Parameters On Identification of WTG Model Parameters and Model Validation Effortsand Model Validation Efforts
Efforts Underway in North AmericaEfforts Underway in North AmericaEfforts Underway in North AmericaEfforts Underway in North America
• NREL Model Parameter Identification Project– Tune parameters to match performance of manufacturer models) using
parameter trajectory sensitivity approach
– Parameters to be published in WECC WGMG guidep g
• IEEE DPWPGMG– Collaboration with several WTG manufacturers and industry stakeholders
in the US and Europein the US and Europe
– Develop specifications and test systems for model validation (underway)
• UWIG/EnerNex– Document generic model validation against measurements
• Hydro QuebecHas done some work in this area– Has done some work in this area
• PSLF and PSSE software developers and consultants
Parameter Sensitivity ApproachParameter Sensitivity ApproachThe output of the simulation and the measured data can be used to find the total error of the measurement Sk
T k( ) dT k( ) k⋅:=Sk
T k( )the total error of the measurement.
Perr = |Pmeas.-Psimulated|
k dk T k( )k
Qerr = |Qmeas.-Qsimulated|
The error and the sensitivity parameter k1 with respect to the error
Actual Wind Plant
p pcan be computed.
Use the other parameters k1, k2, k3, k4 etc Model with the
parameter+
The parameter sensitivity can be observed from the results.
parameter to be tunedInput T(k)-
Parameter kThe trend can be used to drive the changes of the parameters.
Parameter Sensitivity ApproachParameter Sensitivity Approach
Sensitivity of parameter f1 (Sf1) to Perror and Qerror
0.1
0.12
0.06
0.08
Sq_f
1
0.02
0.04
Sp_f
1 an
d Sp_f1Sq_f1
-0.02
00 0.5 1 1.5 2 2.5
-0.04
f1
Parameter Sensitivity ApproachParameter Sensitivity ApproachExample for WECC WT3 Generic Model
• Qualitatively similar results for other outputs.N t th t l t f t h ll d/ l t d i fl
Response of P output Sensitivity of P output to a range of parameters
• Note that a lot of parameters have small and/or correlated influence.• Sensitivities obtained as a by-product of running the simulation.
On Representation of Wind Farms With WECC On Representation of Wind Farms With WECC Generic WTG ModelsGeneric WTG ModelsGeneric WTG ModelsGeneric WTG Models
References:– E. Muljadi, C.P. Butterfield, B. Parsons, A. Ellis, ”Characteristics of
Variable Speed Wind Turbines Under Normal and Fault Conditions”, IEEE PES GM, Tampa, Florida, June 2007., p , ,
– J. Brochu, R. Gagnon, C. Larose, “Validation of the WECC Single‐Machine Equivalent Power Plant”, Presented at the IEEE PSCE DPWPG‐WG Meeting, Seattle, Washington, March 2009.g, , g ,
– WECC Wind Generation Power Flow Modeling Guide
Wind Power Plant RepresentationWind Power Plant RepresentationWind Power Plant RepresentationWind Power Plant Representation
• WECC WGMG recommendation is to use a single‐machine equivalent for power flow and dynamic representation of Wind Power Plants
W
Pad-mounted Transformer Equivalent
Wind Turbine
Collector System
Equivalent
Interconnection Transmission
Line
Station Transformer(s)
POI or connection to the grid Collector System
Station
W Generator Equivalent
PF CorrectionShunt Capacitors-Plant-level
Reactive Compensation
POI or Connection to the Transmission
System
Interconnection Transmission Line …or in special cases (e.g., heterogeneous
feeders or WTGs of different types)…
Feeders and Laterals (overhead and/or underground)
Individual WTGs
~
~ Type 4 WTG
Type 1 WTG
Interconnection Transmission
Line
Station Transformer(s)
Pad-mounted Transformer Equivalent
Collector System
Equivalent
POI or Connection to the Transmission
System PF CorrectionShunt Capacitors
Plant-level Reactive Compensation
Equivalent Collector SystemEquivalent Collector System
• Depends on feeder type (OH/UG) and WPP size
• Zeq and Beq, can be computed from WPP conductor schedule, if available
– For radial feeders with N WTGs and I branches: 2nZ
I
∑ I
21
N
nZjXRZ i
ii
eqeqeq
∑==+= ∑
=
=i
ieq BB1
– Where ni is the number of WTGs connected upstream of the
i-th branch
Thi b i l d il d h– This can be implemented easily on a spreadsheet
Detailed Vs. SingleDetailed Vs. Single‐‐Machine RepresentationMachine RepresentationDetailed Vs. SingleDetailed Vs. Single Machine RepresentationMachine Representation
QWT = 0.435 0 -0.4353-phase fault, all WTGs at 12 m/sec
PP34.5 kV
Q34.5 kV
From « Validation of the WECC Single-Machine Equivalent Power Plant », Presented DPWPG-WG Meeting at IEEE PSCE, March 2009 - Jacques Brochu, Richard Gagnon, Christian Larose, Hydro Quebec
Detailed Vs. SingleDetailed Vs. Single‐‐Machine RepresentationMachine RepresentationDetailed Vs. SingleDetailed Vs. Single Machine RepresentationMachine Representation
0.435 0 -0.435QWT =3-phase fault, different wind speed for each feeder
P
1 and 2 feeders
P34.5 kV
4 feeders = Typical
2 and 4 feeders = Typical
Q34.5 kV
1 feeder
From « Validation of the WECC Single-Machine Equivalent Power Plant », Presented DPWPG-WG Meeting at IEEE PSCE, March 2009 - Jacques Brochu, Richard Gagnon, Christian Larose, Hydro Quebec